Success of Bolus Gastric Feeding After Fundoplication Among Children Who Require Preoperative Jejunal Feeding

Abstract

Purpose:

Management of complicated reflux in infants and children is controversial. Jejunal feedings are used when reflux complications occur with gastric feeds. We sought to determine how successful fundoplication is to allow for return of physiologic gastric feeds in patients requiring jejunal feeds preoperatively.

Methods:

A retrospective review of patients requiring jejunal feeds before fundoplication between 2010 and 2015 was conducted.

Results:

Two hundred thirteen children underwent fundoplication during the study period. One hundred fourteen (49%) children required preoperative jejunal feeds. Median preoperative jejunal feeding trial was 15 days (interquartile range [IQR] 8–36). After fundoplication, gastric feeds were attempted in all patients. Ninety-one (80%) patients tolerated feeds postoperatively without return of preoperative symptoms. Twenty-one (18%) children developed gastric feeding intolerance and were treated with jejunal feeds at a mean of 8 months postoperatively (range 3–17). Ten (9%) children eventually tolerated intragastric bolus feeds, requiring jejunal feeds for a median duration of 2.3 months (IQR 1–5). There were no differences seen in those who were able to tolerate gastric early after the operation and those who did not. Of the patients who were unable to tolerate bolus gastric feeds during the study, a higher proportion had neurologic impairment and were on jejunal feeds for a longer period of time before fundoplication.

Conclusion:

In the majority of patients requiring continuous jejunal feeds to manage complications of reflux, fundoplication allows for transition to gastric bolus feeding.

Background

Over the past several decades, fundoplication procedures have become a common treatment in the surgical management of gastroesophageal reflux disease (GERD) in children.^1–3 While literature reveals a continued level of ambiguity in the definition of reflux disease in children, GERD is well described in patients with feeding difficulties and subsequent failure to thrive.^4,5 As such, anti-reflux surgery is often done in combination with gastrostomy tube placement for patients with feeding intolerance.⁶ Initially, this was described as a protective measure in children, particularly those with neurologic impairment.⁵ There has been evidence to contradict this practice, finding that rates of reflux were not significantly reduced in this population.^7,8

Postpyloric feeding regimens are often implemented in the management of children with complicated reflux, particularly those with issues of aspiration.⁹ Multiple studies have evaluated the efficacy of fundoplication and postpyloric, or gastro-jejunostomy (GJ) feeding tubes in the prevention of complications of reflux.¹⁰ However, there is currently no data on the role of fundoplication in patients requiring jejunal feeds preoperatively. We aimed to study how successful fundoplication is for these patients in allowing them to be managed with gastric feeds.

Materials and Methods

After institutional review board approval, a retrospective chart review was conducted on children who underwent a fundoplication at a single center from July 2010 to July 2015. Patients were identified by current procedural terminology or CPT code 43280. Queried patients were further analyzed to find those who had feeding intolerance requiring jejunal feeds before fundoplication. Demographic data were collected. Charts were reviewed to evaluate for feeding difficulties following fundoplication, specifically those patients who went on to require placement of a GJ tube. The primary objective was to evaluate the incidence of and factors associated with GJ feeding dependence following fundoplication. Perioperative factors were compared in patients requiring and not requiring GJ tube.

Descriptive statistics are reported as means ± standard deviation unless otherwise stated. Categorical data were compared using a chi-square test, and continuous variables with a Student's t-test or Mann–Whitney U-test when appropriate.

Results

Two hundred three patients underwent a laparoscopic fundoplication during a 5-year period. One hundred fourteen patients had postpyloric feeds before a fundoplication. Ninety-five percent of fundoplications were performed laparoscopically, and the majority of patients had a concomitant gastrostomy tube placement.

Sixty-six (58%) patients were male. Median gestational age was 37 weeks (interquartile range [IQR] 31–39), and nearly half the cohort was born preterm. Fifty-eight (51%) patients had neurodevelopmental conditions. Thirty-six (32%) patients had genetic evaluation revealing chromosomal anomalies, and 22 (19%) had seizure disorders requiring anti-epileptics at the time of surgery. Forty-four (39%) patients had a cardiac condition with 25 (21%) children requiring cardiac surgery before their fundoplication. Forty-six (40%) patients had a history of respiratory disease with very few patients requiring oxygen support at the time of surgery.

The most common indication for fundoplication was gastric feeding intolerance. Signs and symptoms of feeding intolerance varied between patients. Emesis was the most common subjective complaint and was observed in 50 (44%) patients. Twenty-four (21%) patients had aspiration events resulting from vomiting spells. Forty (35%) patients had bradycardia, apnea, and desaturation episodes with only 2 patients having clinical findings of cyanosis.

Diagnostic studies were not standardized by our surgical group. Diagnostic workup is performed by our pediatricians and gastroenterology teams and frequently occurred before surgical evaluation. Upper gastrointestinal contrast studies were obtained in 99 (87%) patients and showed reflux in 16 (14%) patients. One hundred nine (95%) patients were treated with gastric acid suppression before fundoplication. The average preoperative nasogastric feeding trial was 1.3 ± 2.5 months. Patients included were those who continued to have symptoms on gastric feeds; all of these patients were trialed on postpyloric feeds with improvement of symptoms preoperatively. The median preoperative jejunal feeding trial was 15 days (IQR 8–36 days).

Median age at surgery was 5 months (IQR 2.6 months–1 year) with a median weight at surgery of 4.8 kg (IQR 3.6–8.3). The median operative time was 87 minutes (IQR 63–103 minutes). Eighty-nine (78%) patients had a concomitant gastrostomy tube placement. There were no intraoperative complications. Gastric feeds were initiated in all patients post-fundoplication. Ninety-one (80%) patients tolerated gastric feeds without return of preoperative symptoms. Twenty-one (18%) patients were unable to tolerate gastric bolus feeds and were treated with jejunal feeds. Twelve (11%) patients had symptomatology identical to the indication of fundoplication, most common of which was emesis with continued aspiration events. Ten of the 21 (48%) patients were eventually able to tolerate bolus intragastric feeds, requiring jejunal feeds for a median duration of 2.3 months (IQR 1–5 months). In those patients requiring jejunal feeds postoperatively, a GJ tube was placed at a mean of 8 months post-fundoplication (IQR 3–17 months) (Table 1).

Table 1.

Comparative Analysis of Preoperative and Operative Variables

	Early gastric feeding success, n = 93	Early jejunal feeding used, n = 21	P
Demographics
Gender: male, n	54	12	1
Age at surgery (months)^*	5 (2.7–12)	3.8 (2–17)	.41^a
Weight (kg)^*	4.9 (3.7–8.2)	4.6 (3.7–8.9)	.76^a
Gestational age (weeks)^*	37 (30–39)	38 (36–39)	.40^a
Preterm <37 weeks, n (%)	43 (46)	9 (43)	.81^b
Comorbidities
Cardiac, n (%)	35 (38)	9 (43)	.81^b
Respiratory, n (%)	37 (40)	9 (43)	.81^b
Seizure disorder, n (%)	19 (20)	3 (14)	.76^b
Neurodevelopmental, n (%)	44 (47)	14 (67)	.15^b
Symptomology
Emesis, n (%)	37 (40)	13 (62)	.09^b
Aspiration events, n (%)	20 (22)	4 (19)	1
Bradycardia/apnea spells, n (%)	35 (38)	5 (24)	.31^b
Feeding trial
Preoperative nasogastric feeding trial (days)	43.3 ± 77.6	46 ± 78.4	.89^a
Preoperative jejunal feeding trial (days)	36.8 ± 66.4	54.8 ± 92	.30^a
Operative
Operative time (minutes)	83.2 ± 31.3	89.6 ± 33.5	.40^a
Nissan fundoplication, n (%)	89 (96)	19 (91)	.30^b
Laparoscopic fundoplication, n (%)	88 (95)	20 (95)	1

Median (interquartile range).

Mann–Whitney U-test.

Chi-squared test.

One patient had a re-do fundoplication. This patient had recurrent symptoms of reflux. Contrast evaluation revealed an intact fundoplication with a hiatal hernia. The hiatal hernia was repaired and the patient's fundoplication was revised. This patient tolerated gastric feeds before recurrence of reflux symptoms. Ten patients presented to the hospital with problems in their GJ tube. These children presented an average of five times to the hospital (range, 1–12). Five patients had issues with clogging, presenting a mean of 1.4 times.^1–3 Six patients had issues with leaking, presenting an average of 2.8 times.^1–10 Eight patients presented with issues of dislodgement or malpositioning an average of 3.3 times.^1–6 One patient died before toleration of feeds.

A comparison of preoperative and operative variables was performed between those patients who ultimately were able to tolerate gastric bolus feeds (n = 103) and those who did not (n = 11). Demographic variables were equivalent between both groups. There was a higher proportion of patients with all neurologic impairments seen in those patients who never tolerated gastric feeds (47% versus 81%; P = .05). However, when this was subdivided into patients with seizure disorders, there were no differences between cohorts. Symptomology did not vary between groups. There were longer preoperative jejunal feeding trials in those patients who were unable to be transitioned to gastric feeds (P = .05). Operative variables were comparable. Comparing those patients who required some jejunal feedings early after fundoplication (21 patients) with those who tolerated gastric feeds immediately (93 patients) revealed no significant differences.

Discussion

In patients with complicated reflux being managed with postpyloric feeds, fundoplication resolved the majority of complications with gastric feeds. Nearly 80% were able to tolerate gastric bolus feeds upon initiation of feeds after fundoplication. There was a small minority of patients who required eventual placement of a GJ tube to resume jejunal feeds. However, approximately half of these patients were able to tolerate bolus feeds within a short interval, less than 6 months, after GJ tube placement. In this review, only 10% of patients had a continued need for postpyloric feeds after fundoplication.

There were little differences between patients who were ultimately able to tolerate bolus feeds and those with continued jejunal feeds. Neurologic impairment was observed in a higher percentage of patients requiring ongoing postpyloric feeds; however, this became less significant when comparing proportions of patients with seizure disorders. Interestingly, preoperative jejunal feeding trials in patients with persistent jejunal feeds seemed to be twice as long as those tolerating gastric feeds. Perhaps the protracted period of stomach rest had a negative impact on the ability of the stomach to subsequently tolerate bolus feeding. There was a large standard deviation for the duration of the preoperative feeding trial, likely resulting from variability in referring practitioner management before surgical involvement.

In patients who received a GJ tube after fundoplication, complications related to the tube occurred in all of them. This resulted in multiple presentations to the hospital and additional procedures. Mechanical complications with jejunal tubes are well recognized.^10,11 Furthermore, jejunal tubes require families to manage continuous feeding plans, which can be arduous. One rationale for GJ tube use, despite these compromises, is the belief that fundoplication will result in equally, if not more, difficult complications. This idea may have been true for the early experience with fundoplication in small children and infants. However, now with randomized data demonstrating that avoiding dissection of the phreno-esophageal membrane decreases the risk of wrap herniation or the need for reoperation, fundoplication is a safer operation.¹² With 1 patient requiring reoperation, this series highlights the feasibility of fundoplication with minimal procedural risk.

Limitations include a retrospective study design with the quality of data relying on the limited nature of a chart review. Without prospectively defined and objective parameters around the diagnosis of complicated reflux leading to postpyloric feeding, the study results can be expected to be influenced by the threshold of referring physicians for placing a postpyloric tube preoperatively. Additionally, the identification of infants suffering from complications of reflux is subjective. At our institution, in patients with feeding intolerance as a result of reflux, criteria for fundoplication is symptom resolution after transitioning from a gastric to a jejunal feeding trial preoperatively. The goal of the operation shifts from simply controlling reflux to the more targeted objective of returning children to a physiologic nutrition plan of gastric bolus feeds.

Conclusions

Laparoscopic fundoplication in infants and children provides the majority of patients the opportunity to return to gastric bolus feeding without the need for continuous jejunal feeds.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Leung

, Wong

, Chung

, Wong

, Tam

. Laparoscopic Nissen fundoplication for gastro-oesophageal reflux disease in infants. Pediatr Surg Int, 2015; 31:93–98.

Kawahara

, Okuyama

, Kubota

, Oue

, Tazuke

, Yagi

, Okada

. Can laparoscopic anti-reflux surgery improve the quality of life in children with neurologic and neuromuscular handicaps. J Pediatr Surg, 2004; 39:1761–1764.

Diaz

, Gibbons

, Heiss

, et al. Anti-reflux surgery outcomes in pediatric gastroesophageal reflux disease. Am J Gastroenterolo, 2005; 100:1844–1852.

Heine

, Reddihough

, Catto-Smith

. Gastro-oesophageal reflux and feeding problems after gastrostomy in children with severe neurological impairment. Dev Med Child Neurol, 1995; 37:320–329.

Jolley

, Smith

, Tunell

. Protective antireflux operation with feeding gastrostomy. Experience with children. Ann Surg, 1985; 201:736–740.

Toporowski-Kowalska

, Gebora-Kowalska

, Fendler

, Popinska

, et al. Diagnosis of gastroesophageal reflux and anti-relux procedures among Polish children with gastrostomies: a 10-year nationwide analysis. Eur J Clin Nutr, 2013; 76:1169–1174.

Ponsky

, Gasior

, Parry

, Sharp

, Boulanger

, Parry

, et al. Need for subsequent fundoplication after gastrostomy based on patient characteristics. J Surg Res, 2013; 179:1–4.

Barnhart

, Hall

, Mahant

, et al. Effectiveness of fundoplication at the time of gastrostomy in infants with neurological impairment. JAMA Pediatr, 2013; 167:911–918.

Capriati

, Cardile

, Chiusolo

, et al. Clinical management of post-pyloric enteral feeding in children. Expert Rev Gastroenterol Hepatol, 2015; 9:929–941.

10.

Stone

, Hester

, Daniel

, et al. Effectiveness of fundiplication or gastrojejunal feeding in children with neurologic impairment. Hosp Pediatr, 2017; 7:140–148.

11.

Godbole

, Margabanthu

, Crabbe

, et al. Limitations and uses of gastrojejunal feeding tubes. Arch Dis Child, 2002; 86:134–137.

12.

St. Peter

, Barnhart

, Ostlie

, et al. Minimal vs extensive esophageal mobilization during laparoscopic fundoplication: A prospective randomized trial. J Pediatr Surg, 2011; 46:163–168.